gms | German Medical Science

Background: Chest pain is a frequent complaint in primary care practice. In a minority of only 10–15% of cases it is caused by coronary heart disease (CHD). However, general practitioners (GPs) must diagnose these patients accurately and refer if needed. The guideline has as its objective to support GPs’ decision making in this area. The emphasis is on history taking and physical examination.

Methods: We designed a research programme to inform guideline development. This included a qualitative study to explore actual reasoning of GPs regarding chest pain and the identification of serious illness; three cross-sectional studies of patients with chest pain in primary care (overall n=2500 patients); a conventional systematic review and one based on individual patient data; a practice test with qualitative and quantitative methods.

Results: Our research programme in cooperation with national and international partners improved the evidence base of recommendations considerably. A simple decision tool, the Marburg Heart Score, was developed and validated in two independent samples. Several traditional items of the history were shown not to discriminate between CHD and less serious aetiologies.

Conclusion: Clinical diagnosis is topic of considerable complexity for guideline developers. Often the evidence base is poor. A focused research programme can inform guideline development. However, this requires sufficient resources including time.